US7536528B2 - Memory arrangement - Google Patents
Memory arrangement Download PDFInfo
- Publication number
- US7536528B2 US7536528B2 US11/679,609 US67960907A US7536528B2 US 7536528 B2 US7536528 B2 US 7536528B2 US 67960907 A US67960907 A US 67960907A US 7536528 B2 US7536528 B2 US 7536528B2
- Authority
- US
- United States
- Prior art keywords
- memory
- data
- bank access
- bank
- interface
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Expired - Fee Related, expires
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F12/00—Accessing, addressing or allocating within memory systems or architectures
- G06F12/02—Addressing or allocation; Relocation
- G06F12/06—Addressing a physical block of locations, e.g. base addressing, module addressing, memory dedication
- G06F12/0607—Interleaved addressing
Definitions
- the memory arrangement is not driven via parallel application of data signals, address signals and control signals, as in typical conventional memory arrangements, but rather via data packets that are transmitted at high speed and at high frequency, in accordance with a predefined protocol, between, for example, a computer system and an interface of the memory arrangement.
- These data packets may contain write data, read data, addressing data, and command data.
- the received data are interpreted and processed as commands, address data, or write data.
- read data can also be coded into data packets by the interface and transmitted to, for example, a computer system.
- the memory arrangement of such a type comprises several memory banks that can be accessed via several memory-bank access devices and that each include a plurality of memory cells.
- One embodiment provides a memory arrangement including an interface configured to transmit, code and/or decode data in the form of data packets in accordance with a predefined protocol.
- the memory arrangement includes at least two memory banks, each memory bank including at least one memory cell.
- the memory arrangement includes at least two memory-bank access devices configured to facilitate accessing the data of the at least one memory cell of each of the at least two memory banks.
- the memory arrangement includes at least two temporary storage devices configured to temporarily store data being transmitted between the interface and the at least two memory-bank access devices. Each of the at least two temporary storage devices is connected to the interface and to one of the at least two memory-bank access devices.
- FIG. 1 is a schematic top view of a memory arrangement according to one embodiment with two temporary storage devices.
- FIG. 2 illustrates a schematic top view of one embodiment of the memory arrangement with four temporary storage devices.
- FIG. 3 is a schematic top view of a memory arrangement in which a data flow is represented
- FIG. 4 is a schematic top view of a memory arrangement with only one temporary storage device.
- FIG. 3 illustrates an example memory arrangement 10 with sixteen memory banks 6 which are arranged in a four-by-four matrix.
- the upper eight memory banks 6 are spaced from the lower eight memory banks 6 , so that a strip-shaped area 7 , which is also referred to as spine, is formed between the upper eight memory banks 6 and the lower eight memory banks 6 .
- the sixteen memory banks 6 are further partitioned into two groups, each of eight memory banks 6 , in which, referring to FIG.
- a left group defines the eight memory banks 6 located on the left in a top view
- a right group defines the eight memory banks 6 located on the right in the top view.
- a memory bank 6 pertaining to the left group is accessed via the first memory-bank access device 1
- a memory bank 6 pertaining to the right group is accessed via the second memory-bank access device 2 .
- the interface 4 , the memory-bank access devices 1 and 2 , and further devices of the memory arrangement, such as, for example, a clock-pulse supply, are arranged in the spine 7 .
- the interface 4 is arranged substantially at a center in the longitudinal direction of the spine 7 , the first memory-bank access device 1 , which is assigned to the left group of memory banks, is substantially arranged at a center in a left half of the spine 7 , and the second memory-bank access device 2 , which is assigned to the right group of memory banks, is substantially arranged at a center in a right half of the spine 7 .
- Write data received by the interface 4 are transmitted in the direction of arrows 8 and 9 to the corresponding memory-bank access devices 1 and 2 , respectively.
- read data are transmitted contrary to the arrow directions of the arrows 8 and 9 from the memory-bank access devices 1 and 2 , respectively, to the interface 4 .
- An arrangement of such a type enables a uniform timing response of the individual memory banks to be obtained.
- temporary storage devices are inserted between the interface and the memory-bank access devices.
- these temporary storage devices cause the write data that are decoded from a data packet to be firstly stored by the interface in the temporary storage device, in order then to be stored in the memory banks in temporally independent manner via the memory-bank access devices.
- write data can be transmitted to the memory arrangement at a speed that is far higher than the maximum possible speed of storage of these write data in the memory banks.
- a command for making read data available can be received by the interface, and the read operation can be triggered, so that the interface is then ready for receiving further write data, read data, addressing data, or command data, while the read data from the memory banks are sent to the temporary storage devices via the memory-bank access devices, in order to be read out from there at a later time via the interface.
- a further advantage of the temporary storage is the on-chip timing response can be decoupled from the off-chip timing response.
- FIG. 4 illustrates an example memory arrangement 10 with a temporary storage device 3 which is suitable to temporarily store data from data packets that are received by the interface 4 .
- the interface 4 is connected to the temporary storage device 3 via connecting lines in an area 11
- the temporary storage device 3 is in turn connected to the memory-bank access devices 1 and 2 via connecting lines in areas 12 .
- a symmetrical arrangement of both the temporary storage device 3 and the interface 4 is employed for reasons concerning the timing.
- FIG. 4 illustrates that a central arrangement of the temporary storage device 3 in the center of the spine 7 has the consequence that, on the one hand, a large chip area is employed in the center of the spine 7 , which as a result is no longer available to other central functions of the memory arrangement, and that, on the other hand, relatively large areas 12 employed within the spine 7 for the connecting lines from the temporary storage device 3 to the memory-bank access devices 1 and 2 .
- Located in the areas 12 are, for example, 72 lines in each instance, which correspond to 64 lines for 64 data bits of the data-bank memory width plus eight lines for eight masking bits.
- a further advantage of the use of the temporary storage device is that, for example, the temporary storage device temporarily stores the contents of several related packets of write data, and these contents are then stored in the memory banks in one write access via the memory-bank access devices. If, for example, 64 bits can be written simultaneously via a memory-bank access device, and two memory-bank access devices are present, it is possible to transmit, firstly, 128 data bits and additional 16 masking bits which can each be used for the purpose of masking eight of the 128 data bits in each instance in, for example, four data packets, and then to transmit these 128 data bits and 16 masking bits in one write operation to the memory-bank access devices which then store the unmasked bits of the 128 data bits in the memory banks. As a result, the system performance is further enhanced.
- One embodiment of a memory arrangement comprises an interface, which is suitable for transmitting, coding and/or decoding data in the form of data packets in accordance with a predefined protocol, at least two memory banks, which each include at least one memory cell, at least two memory-bank access devices, via which the data of the memory cells of the at least two memory banks are accessed, and at least two temporary storage devices, the temporary storage devices being suitable for temporarily storing data that are being transmitted between the interface and the at least two memory-bank access devices, and each of the at least two temporary storage devices being connected both to the interface and to one of the at least two memory-bank access devices.
- the memory arrangement includes an even number of memory banks, a first memory-bank access device and a second memory-bank access device.
- the memory cells pertaining to a first half of the memory banks are accessed via the first memory-bank access device, and the memory cells pertaining to a second half of the memory banks are accessed via the second memory-bank access device.
- two temporary storage devices and two memory-bank access devices can be arranged on the memory arrangement, one of the two temporary storage devices being arranged in each instance adjacent to one of the two memory-bank access devices.
- each of the two temporary storage devices only that portion of the data is temporarily stored which is being transmitted via the adjacently arranged memory-bank access device.
- the chip area of each individual temporary storage device that is employed amounts approximately to one half of the chip area of a comparable unpartitioned temporary storage device.
- two memory-bank access devices and four temporary storage devices are provided on the memory arrangement, two of the four temporary storage devices being arranged in each instance adjacent to each of the two memory-bank access devices.
- each of the four temporary storage devices only that portion of the data is temporarily stored in each instance which is being transmitted via the adjacently arranged memory-bank access devices, so that the chip area of each of the four temporary storage devices corresponds approximately to one quarter of a comparable unpartitioned temporary storage device.
- the interface of the memory arrangement is configured in such a manner that the interface is able to transmit code and/or decode write data, read data, addressing data, and command data in the form of data packets.
- the data packets may be structured in such a manner that a data packet includes a defined number of bits which are transmitted via several parallel lines.
- the number of parallel lines may be distinctly less than the number of bits of a data packet, by several consecutive bit groups being transmitted via the parallel lines and then being assembled into a data packet.
- Data packets with a length of, for example, 54 bits can consequently, for example, be transmitted via six lines in nine consecutive 6 bit groups.
- the interface synchronizes to the serial data stream and parallelizes the serially received data.
- Data that are being transmitted from the memory arrangement to, for example, a computer system are serialized by the interface.
- the memory arrangement includes several memory banks that can be accessed via several memory-bank access devices.
- memory-bank access devices it holds that memory cells of one or more memory banks can be accessed via a memory-bank access device.
- Access to memory cells of a defined memory bank is possible only from exactly one memory-bank access device. Consequently several memory banks can be assigned to one memory-bank access device, but one memory bank cannot be assigned to various memory-bank access devices.
- the memory arrangement is employed in dynamic random access memory (DRAM) memory modules on a common chip.
- DRAM dynamic random access memory
- ROM read only memory
- the memory arrangement is employed in a computer system as read/write memory or DRAM memory.
- FIG. 1 illustrates schematically an embodiment of a memory arrangement 10 with an even number of memory banks 6 , a first memory-bank access device 1 , and a second memory-bank access device 2 .
- the sixteen memory banks 6 are partitioned into two areas. Like in FIG. 3 , eight memory banks 6 are located in an upper area, and eight memory banks 6 are located in a lower area. Located between the upper area and the lower area is a narrow, strip-shaped area 7 in which the memory-bank access devices 1 and 2 and an interface 4 are located. This narrow area 7 between each of the eight memory banks is designated as the spine 7 .
- the sixteen memory banks 6 are further partitioned into two groups of, in each instance, eight memory banks 6 , in which, referring to FIG.
- a left group defines the eight memory banks located on the left in a top view
- a right group defines the eight memory banks located on the right in the top view.
- One of the memory banks 6 pertaining to the left group is accessed via the first memory-bank access device 1
- one of the memory banks 6 pertaining to the right group is accessed via the second memory-bank access device 2 .
- the interface 4 is arranged substantially at a center in a longitudinal direction of the spine 7 ; the first memory-bank access device 1 , which is assigned to the left group of memory banks, is arranged substantially at a center of a left half of the spine 7 ; and the second memory-bank access device 2 , which is assigned to the right group of memory banks, is arranged substantially at a center of a right half of the spine 7 .
- a first temporary storage device 5 and a second temporary storage device 5 are arranged respectively adjacent to one of the memory-bank access devices 1 and 2 , respectively, the first temporary storage device 5 being arranged in an area that extends from the center of the spine in the direction of the first memory-bank access device 1 , and the second temporary storage device 5 being arranged in an area that extends from the center of the spine in the direction of the second memory-bank access device 2 .
- the temporary storage devices 5 are arranged in the aforementioned areas in such a manner that they each have a spacing from the respectively adjacently arranged memory-bank access devices 1 and 2 that is as small as possible.
- the two temporary storage devices 5 are connected to the interface 4 via connecting lines that are arranged in an area 11 .
- the two memory-bank access devices 1 and 2 are each connected to one of the two temporary storage devices 5 via connecting lines that are arranged in areas 12 .
- each temporary storage device 5 corresponds substantially to one half of the area requirement of the unpartitioned temporary storage device 3 of FIG. 4 , since in each temporary storage device 5 only those data are stored which are to be stored in the memory banks 6 via the memory-bank access device 1 or 2 to which the respective temporary storage device 5 is connected. Since the write data are received by the interface 4 in several data packets, after they have been decoded by the interface 4 the write data are transmitted packet-by-packet from the interface 4 to the temporary storage devices 5 . As a result of this partly serialized transmission between the interface 4 and the temporary storage device 5 , the number of necessary connecting lines in the area 11 between the interface 4 and the temporary storage devices 5 is reduced.
- each portion of the area 11 approximately 30 connecting lines are arranged, wherein the portions of area 11 extend from the center of the spine in the direction of the memory-bank access devices 1 and 2 , respectively.
- 72 lines are arranged, wherein the areas 12 likewise extend from the center of the spine in the direction of the memory-bank access devices 1 and 2 , respectively.
- FIG. 2 illustrates a schematic top view of an embodiment of a memory arrangement 10 .
- the interface 4 , the memory banks 6 , the spine 7 and the memory-bank access devices 1 , 2 are arranged similar to as described above in the description of FIG. 1 .
- four temporary storage devices 5 are arranged at four places in the memory arrangement 10 , specifically in such a manner that two temporary storage devices 5 are arranged adjacent to each memory-bank access device 1 and 2 , as illustrated in FIG.
- a first of the temporary storage devices 5 being arranged to the left of the first memory-bank access device 1
- a second of the temporary storage devices 5 being arranged between the center of the spine 7 and the first memory-bank access device 1
- a third of the temporary storage devices 5 being arranged between the center of the spine 7 and the second memory-bank access device 2
- a fourth of the temporary storage devices 5 being arranged to the right of the second memory-bank access device 2 .
- the chip area employed for each of the four temporary storage devices 5 corresponds, in each instance, to approximately one quarter of the surface area of the unpartitioned temporary storage device 3 illustrated in FIG. 4 , because in each of the four temporary storage devices 5 only that portion of the data is stored which is to be stored in the memory banks 6 via the memory-bank access device 1 or 2 to which the respective temporary storage devices 5 are connected. In turn, approximately 30 connecting lines are located in the area 11 between the interface 4 and the temporary storage devices 5 . In the areas 12 , on the other hand, only 36 connecting lines are now arranged between, in each instance, a temporary storage device and a memory-bank access device.
- temporary storage devices embodiments described above may be employed advantageously in conjunction with the partitioning, according embodiments, of the temporary storage devices also in the case of the readout of data from the memory banks 6 .
- temporary storage devices 5 that are arranged between memory-bank access devices 1 and 2 and an interface 4 , which is configured for transmitting, coding and decoding read data, addressing data and command data, have read data written to them from the memory banks 6 via the memory-bank access devices 1 and 2 by reason of a read request.
- a readout of the temporary storage devices 5 is possible in temporally independent manner via data packets that are coded and transmitted by the interface 4 .
- temporary storage devices temporarily store commands and command sequences that are executed in a manner temporally independent of the receiving of the commands or command sequences, in order, for example, to refresh memory banks or to provide read data.
Abstract
Description
Claims (25)
Applications Claiming Priority (2)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
DE102006009027.6 | 2006-02-27 | ||
DE102006009027A DE102006009027A1 (en) | 2006-02-27 | 2006-02-27 | Memory arrangement comprises interface, two memory banks with one memory cell, and two memory bank access devices, intermediate storage devices are provided for intermediate saving of data |
Publications (2)
Publication Number | Publication Date |
---|---|
US20070204116A1 US20070204116A1 (en) | 2007-08-30 |
US7536528B2 true US7536528B2 (en) | 2009-05-19 |
Family
ID=38319888
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
US11/679,609 Expired - Fee Related US7536528B2 (en) | 2006-02-27 | 2007-02-27 | Memory arrangement |
Country Status (2)
Country | Link |
---|---|
US (1) | US7536528B2 (en) |
DE (1) | DE102006009027A1 (en) |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US10089043B2 (en) | 2013-03-15 | 2018-10-02 | Micron Technology, Inc. | Apparatus and methods for a distributed memory system including memory nodes |
Families Citing this family (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US9779057B2 (en) | 2009-09-11 | 2017-10-03 | Micron Technology, Inc. | Autonomous memory architecture |
US9779138B2 (en) | 2013-08-13 | 2017-10-03 | Micron Technology, Inc. | Methods and systems for autonomous memory searching |
US10003675B2 (en) | 2013-12-02 | 2018-06-19 | Micron Technology, Inc. | Packet processor receiving packets containing instructions, data, and starting location and generating packets containing instructions and data |
Citations (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US5898623A (en) | 1997-10-09 | 1999-04-27 | International Business Machines Corporation | Input port switching protocol for a random access memory |
US6501698B1 (en) * | 2000-11-01 | 2002-12-31 | Enhanced Memory Systems, Inc. | Structure and method for hiding DRAM cycle time behind a burst access |
US20040202027A1 (en) | 2003-01-21 | 2004-10-14 | Maksim Kuzmenka | Buffer amplifier architecture for semiconductor memory circuits |
US20050097264A1 (en) | 2003-10-31 | 2005-05-05 | Park Kwang-Soo | Memory module with registers |
US20050180255A1 (en) | 1997-10-10 | 2005-08-18 | Tsern Ely K. | Memory device having a read pipeline and a delay locked loop |
US7308524B2 (en) * | 2003-01-13 | 2007-12-11 | Silicon Pipe, Inc | Memory chain |
US7418526B2 (en) * | 2004-03-29 | 2008-08-26 | Micron Technology, Inc. | Memory hub and method for providing memory sequencing hints |
-
2006
- 2006-02-27 DE DE102006009027A patent/DE102006009027A1/en not_active Ceased
-
2007
- 2007-02-27 US US11/679,609 patent/US7536528B2/en not_active Expired - Fee Related
Patent Citations (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US5898623A (en) | 1997-10-09 | 1999-04-27 | International Business Machines Corporation | Input port switching protocol for a random access memory |
US20050180255A1 (en) | 1997-10-10 | 2005-08-18 | Tsern Ely K. | Memory device having a read pipeline and a delay locked loop |
US6501698B1 (en) * | 2000-11-01 | 2002-12-31 | Enhanced Memory Systems, Inc. | Structure and method for hiding DRAM cycle time behind a burst access |
US7308524B2 (en) * | 2003-01-13 | 2007-12-11 | Silicon Pipe, Inc | Memory chain |
US20040202027A1 (en) | 2003-01-21 | 2004-10-14 | Maksim Kuzmenka | Buffer amplifier architecture for semiconductor memory circuits |
US20050097264A1 (en) | 2003-10-31 | 2005-05-05 | Park Kwang-Soo | Memory module with registers |
US7418526B2 (en) * | 2004-03-29 | 2008-08-26 | Micron Technology, Inc. | Memory hub and method for providing memory sequencing hints |
Non-Patent Citations (1)
Title |
---|
Crisp, Richard, "High Bandwidth RDRAM Technology Reduces System Cost," IEEE Proceedings of COMPCON '96, pp. 365-377 (1996). |
Cited By (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US10089043B2 (en) | 2013-03-15 | 2018-10-02 | Micron Technology, Inc. | Apparatus and methods for a distributed memory system including memory nodes |
US10761781B2 (en) | 2013-03-15 | 2020-09-01 | Micron Technology, Inc. | Apparatus and methods for a distributed memory system including memory nodes |
Also Published As
Publication number | Publication date |
---|---|
US20070204116A1 (en) | 2007-08-30 |
DE102006009027A1 (en) | 2007-08-30 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
US6963949B2 (en) | Pipelined packet-oriented memory system having a unidirectional command and address bus and a bidirectional data bus | |
KR101467623B1 (en) | Independently controlled virtual memory devices in memory modules | |
US7966446B2 (en) | Memory system and method having point-to-point link | |
US6535450B1 (en) | Method for selecting one or a bank of memory devices | |
US20060050591A1 (en) | Address coding method and address decoder for reducing sensing noise during refresh operation of memory device | |
US20090097348A1 (en) | Integrated circuit including a memory module having a plurality of memory banks | |
US7536528B2 (en) | Memory arrangement | |
US20090106504A1 (en) | Memory system and method for operating a memory system | |
US4951246A (en) | Nibble-mode dram solid state storage device | |
CN102541769A (en) | Memory interface access control method and device | |
US6363017B2 (en) | Method and apparatus for alternate operation of a random access memory in single-memory operating mode and in combined multi-memory operating mode | |
US8539196B2 (en) | Hierarchical organization of large memory blocks | |
US7523250B2 (en) | Semiconductor memory system and semiconductor memory chip | |
US7725647B2 (en) | Memory arrangement having efficient arrangement of devices | |
US8205021B2 (en) | Memory system and integrated management method for plurality of DMA channels | |
KR100314129B1 (en) | Semiconductor implementing bank and data input/output line architecture to reduce data input/output line loading | |
EP4086908B1 (en) | Data processing circuit, and device | |
US7729198B2 (en) | Synchronous memory circuit | |
CN1588335A (en) | Storage access device and relative device | |
US20060095652A1 (en) | Memory device and method for receiving instruction data | |
CN115116512A (en) | Data processing circuit and device | |
CN115116513A (en) | Data processing circuit and device | |
KR100754361B1 (en) | Multi-port memory device enabling share of dq power | |
US20050144373A1 (en) | System and method for variable array architecture for memories | |
US20070033351A1 (en) | Semiconductor memory module unit for point-to-point data interchange |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
AS | Assignment |
Owner name: QIMONDA AG, GERMANY Free format text: ASSIGNMENT OF ASSIGNORS INTEREST;ASSIGNORS:WALLNER, PAUL;DUDHA, CHAITHANYA;REEL/FRAME:019289/0835 Effective date: 20070426 |
|
FEPP | Fee payment procedure |
Free format text: PAYOR NUMBER ASSIGNED (ORIGINAL EVENT CODE: ASPN); ENTITY STATUS OF PATENT OWNER: LARGE ENTITY |
|
FPAY | Fee payment |
Year of fee payment: 4 |
|
AS | Assignment |
Owner name: INFINEON TECHNOLOGIES AG, GERMANY Free format text: ASSIGNMENT OF ASSIGNORS INTEREST;ASSIGNOR:QIMONDA AG;REEL/FRAME:035623/0001 Effective date: 20141009 |
|
AS | Assignment |
Owner name: POLARIS INNOVATIONS LIMITED, IRELAND Free format text: ASSIGNMENT OF ASSIGNORS INTEREST;ASSIGNOR:INFINEON TECHNOLOGIES AG;REEL/FRAME:037147/0487 Effective date: 20150708 |
|
AS | Assignment |
Owner name: POLARIS INNOVATIONS LIMITED, IRELAND Free format text: ASSIGNMENT OF ASSIGNORS INTEREST;ASSIGNOR:INFINEON TECHNOLOGIES AG;REEL/FRAME:037171/0719 Effective date: 20150708 |
|
REMI | Maintenance fee reminder mailed | ||
LAPS | Lapse for failure to pay maintenance fees | ||
STCH | Information on status: patent discontinuation |
Free format text: PATENT EXPIRED DUE TO NONPAYMENT OF MAINTENANCE FEES UNDER 37 CFR 1.362 |
|
FP | Lapsed due to failure to pay maintenance fee |
Effective date: 20170519 |